Regular expressions with validations in RoR 4 - ruby-on-rails

Question

There is the following code:
class Product < ActiveRecord::Base
validates :title, :description, :image_url, presence: true
validates :price, numericality: {greater_than_or_equal_to: 0.01}
validates :title, uniqueness: true
validates :image_url, allow_blank: true, format: {
with: %r{\.(gif|jpg|png)$}i,
message: 'URL must point to GIT/JPG/PNG pictures'
}
end
It works, but when I try to test it using "rake test" I'll catch this message:
rake aborted!
The provided regular expression is using multiline anchors (^ or $), which may present a security risk. Did you mean to use \A and \z, or forgot to add the :multiline => true option?
What does it mean? How can I fix it?

Answer 1

^ and $ are Start of Line and End of Line anchors. While \A and \z are Permanent Start of String and End of String anchors.
See the difference:
string = "abcde\nzzzz"
# => "abcde\nzzzz"
/^abcde$/ === string
# => true
/\Aabcde\z/ === string
# => false
So Rails is telling you, "Are you sure you want to use ^ and $? Don't you want to use \A and \z instead?"
There is more on the rails security concern that generates this warning here.

Answer 2

This warning raises because your validation rule is vulnerable for javascript injection.
In your case \.(gif|jpg|png)$ matches till the end of the line. So your rule will validate this value pic.png\nalert(1); as true:
"test.png\n<script>alert(1)</script>" === /\.(gif|jpg|png)$/i
# => true
"test.png\n<script>alert(1)</script>" === /\.(gif|jpg|png)\z/i
# => false
Read the acticles:
http://caiustheory.com/validating-data-with-regular-expressions-in-ruby
http://guides.rubyonrails.org/security.html#regular-expressions

Answer 3

The problem regexp is not in devise, but rather lives in config/initializers/devise.rb. Change:
# Regex to use to validate the email address
config.email_regexp = /^([\w\.%\+\-]+)#([\w\-]+\.)+([\w]{2,})$/i
to:
# Regex to use to validate the email address
config.email_regexp = /\A([\w\.%\+\-]+)#([\w\-]+\.)+([\w]{2,})\Z/i

Answer 4

The warning is telling you that strings like the following will pass validation, but it is probably not what you want:
test = "image.gif\nthis is not an image"
re = /\.(gif|jpg|png)$/i
re.match(test) #=> #<MatchData ".gif" 1:"gif">
Both ^ and $ matches the start/end of any line, not the start/end of the string. \A and \z matches the start and the end of the full string, respectively.
re = /\.(gif|jpg|png)\z/i
re.match(test) #=> nil
The second part of the warning (“or forgot to add the :multiline => true option”) is telling you that if you actually want the behaviour of ^ and $ you can simply silence the warning passing the :multiline option.

Answer 5

If Ruby wants to see \z instead of the $ symbol sign, for security, you need to give it to him, then the code would look like this :
validates :image_url, allow_blank: true, format: {with: %r{\.(gif|jpg|png)\z}i, message: 'URL must point to GIF, JPG, PNG.'}